Maternal exposure to concentrations of inorganic arsenic (iAs) in excess of 10 parts per billion (ppb) has been identified as a risk factor for clinically recognized spontaneous pregnancy loss in several epidemiologic studies. The majority of the studies reported to date assessed persons residing in regions well-known for substantial levels of groundwater iAs contamination. A far larger global population, likely to include tens of millions of U.S. residents, are exposed to mild-moderate iAs concentrations via drinking water contamination; those less than 10 ppb iAs. Although limited, human, animal and in vitro studies suggest an increased risk for spontaneous pregnancy loss in association with exposure to mild-moderate drinking water iAs concentrations. To date, only one study, conducted from 1976-1978, targeted a population with drinking water iAs exposures less than 10 ppb, and detected a 50% increased risk for spontaneous loss of a clinically recognized pregnancy. Given the high baseline risk for the spontaneous loss of clinically recognized pregnancies among humans (i.e., 12-24%), and the widespread distribution of exposure to drinking water with mild-moderate iAs contamination even a modest increase in risk would incur a substantial number of attributable events. We propose a preliminary evaluation of associations between drinking water iAs exposure and the spontaneous loss of clinically recognized pregnancies of less than 20 weeks completed gestation, and to assess the feasibility of participant recruitment and data collection strategies at a large urban medical center in western Romania. This area of Romania is well-known for mild-moderate iAs contamination of drinking water sources providing a wide range of exposures below 10 ppb. Moreover, prenatal care is compulsory and government funded in Romania facilitating nearly complete ascertainment of the proposed study population comprised of newly pregnant women 18-44 years of age. We will employ a case-control study design with incidence density sampling of 150 cases and 150 controls matched by gestational age (1:1). Individual historical drinking water exposure data collected by nurse-administered questionnaire, augmented by laboratory iAs measures in reported drinking water sources will be used to assign exposure across critical windows for reproduction. We hypothesize that exposures to greater amounts of iAs through the consumption of drinking water at home and at the workplace, prior to and during gestation, will increase the risk for spontaneous pregnancy loss. Our proposed study will provide preliminary data to permit and evaluation of the best approach for a larger multi- site confirmatory study to address this critical data gap concerning low-moderate drinking water iAs contamination and spontaneous pregnancy loss. These data are vital for the protection of expectant mothers and their embryos and fetuses from a possibly lethal and widespread risk factor for loss.